Biology subjects

Ralph, M. L.

Publications and source records attributed to Ralph, M. L..

4 recordsLinked to original sources

bioRxiv2026

Semantic distance differently modulates FPVS-EEG responses to words and pictures

The organization of semantic knowledge within the brain has long been studied through various theoretical frameworks and is still a matter of debate. Many studies that have helped advancing our knowledge about this topic have approached it in the context of its deterioration, notably in patients with semantic dementia. In healthy subjects, the question whether and how semantic distance between concepts represented by picture or word stimuli influences brain responses remains understudied. In this study with 24 healthy subjects, we used electroencephalography (EEG) recordings coupled with a fast periodic visual stimulation (FPVS) approach, for the first time, to assess the semantic distance effect with different stimulus modalities. Picture or word stimuli from a reference category (birds) appeared every fourth item among a 4-Hz stream of either man-made objects (forming the high distance [HD] condition) or other animals (low distance [LD] condition). Within a few minutes of recording, EEG responses were observed at the pre-determined frequency of the reference category presentation (at 1 Hz), suggesting its exemplars were activated at least sufficiently to be automatically discriminated from different superordinate (HD) and basic (LD) categories, for both pictures and words. Pictures and words elicited a different pattern of response to semantic distance, with larger amplitudes for HD than LD conditions for pictures but a reversed pattern of somewhat greater amplitudes for LD than HD conditions for words. Our findings support a similarity-based structure of semantic representations and provide evidence for a differential mapping between semantic and picture/word surface representations.

neuroscience↗

bioRxiv2026

A lateral temporal network for transmodal combinatorial semantics: Convergent evidence from a multi-study investigation

This study integrates three literatures typically examined in isolation: single-concept semantics, combinatorial semantics, and theory of mind (ToM). We argue that these domains share overlapping computational principles and neuroanatomical networks. Here, we report three major investigations with converging methodological approaches: a meta-analysis of 410 neuroimaging studies, a large omnibus fMRI cross-study comparison (drawing on data from over 150 participants), and two targeted fMRI studies that integrate large language models with traditional psycholinguistic measures, allowing us to quantify combinatorial processing demands and predict brain activation. Across methods, convergent evidence identified a stable, bilateral ATL-STS-TPJ network supporting transmodal combinatorial semantic processing. In addition, the anterior temporal lobe (ATL) showed graded functional specialisation: ventral ATL responded equivalently to single-concept and combinatorial semantics, consistent with a domain-general semantic role, whereas lateral superior ATL was selectively recruited by combinatorial demands. Although ATL-STS-TPJ overlapped with ToM-related activation, targeted control analyses demonstrated that this overlap was eliminated when lexico-syntactic and semantic coherence demands were controlled. Together, these findings support a multimodal combinatorial semantic network centred on bilateral ATL-STS-TPJ with implications for theories of semantic and social cognition. On the basis of these results, we propose a unified theoretical framework of semantic processing.

neuroscience↗

bioRxiv2024

Are resting-state networks the brain's cognitive atoms? Differential dynamic reconfiguration of functional brain networks across tasks and at rest

It is increasingly popular to utilise functional connectivity (FC) analyses of resting-state functional magnetic resonance imaging (fMRI) to characterize human functional brain networks and to use the emergent resting-state networks (RSNs) in basic and clinical neuroscience. Often, they are treated as atomic building blocks that underpin human cognition. However, the true function of these RSNs, as well as the relationship between intrinsic and task-evoked functional brain networks, are complex and incompletely characterized. Here, we investigated the functional characteristics of the intrinsic and extrinsic networks using resting-state and task fMRI. Independent component analysis (ICA) was used to estimate spatiotemporal functional networks during tasks and at rest, and to compare the spatiotemporal properties of each network. While there was some spatial correspondence between the RSNs and task-evoked networks, our results demonstrated that the task-evoked functional networks were different from the RSNs in task-relatedness as well as spatial topology. Furthermore, the degree of topological differences between the RSNs and task-evoked networks was modulated by a given task. Comparison between the RSNs and task-evoked networks showed that tasks reconfigure the RSNs by changing FC with various brain regions specific to the task condition. Our findings indicate that the brain does not maintain an "invariant intrinsic" network architecture when it engages in a task. Instead, the tasks reconfigure the network architectures, thereby accommodating specific computational/representational task requirements through flexible interactions between demand-specific regions. Thus, the results suggest that task fMRI is required to understand the full repertoire of the brains functional architecture. Significant StatementResting-state networks (RSNs) could offer a critical foundation for understanding the brains intrinsic organization. However, the functional nature of these intrinsic networks, and their relationship to those activated during specific tasks, are complex and not fully understood. We undertook a comprehensive examination of the functional attributes of intrinsic and extrinsic networks. Although we observed some spatial congruence between RSNs and task-evoked networks, we found fundamental differences in their task-relatedness and spatial topology. These findings highlight the dynamic nature of the brains functional networks, which adapt to specific task demands through flexible interactions among task-specific regions. Thus, task fMRI is essential for a comprehensive understanding of how the human brain dynamically reconfigures its functional architecture in response to external demands, providing valuable insights into the cognition and behaviour.

neuroscience↗

bioRxiv2024

The dimensionality of neural coding for cognitive control is gradually transformed within the lateral prefrontal cortex

Implementing cognitive control relies on neural representations that are inherently high-dimensional and distribute across multiple subregions in the prefrontal cortex (PFC). Traditional approaches tackle prefrontal representations by reducing them into a unidimensional measure (univariate amplitude) or using them to distinguish a limited number of alternatives (pattern classification). By contrast, representational similarity analysis (RSA) enables flexibly formulating various hypotheses about informational contents underlying the neural codes, explicitly comparing hypotheses, and examining the representational alignment between brain regions. Here, we used a multifaceted paradigm wherein the difficulty of cognitive control was manipulated separately for five cognitive tasks. We used RSA to unveil representational contents, measure the representational alignment between regions, and quantify representational generality vs. specificity. We found a graded transition in the lateral PFC: The dorsocaudal PFC was tuned to the information about behavioural effort, preferentially connected with the parietal cortex, and representationally generalisable across domains. The ventrorostral PFC was tuned to the abstract structure of tasks, preferentially connected with the temporal cortex, and representationally specific. The middle PFC (interposed between dorsocaudal and ventrorostral PFC) was tuned to individual task-sets, ranked in the middle in terms of connectivity and generalisability. Furthermore, whether a region was dimensionally rich or thin co-varied with its functional profile: Low dimensionality (only gist) in the dorsocaudal PFC dovetailed with better generality, whereas high dimensionality (gist plus details) in the ventrorostral PFC corresponded with better ability to encode subtleties. Our findings, collectively, demonstrate how cognitive control is decomposed into distinct facets that transition steadily along prefrontal subregions. SignificanceCognitive control is known to be a high-dimensional construct, implemented along the dorsocaudal-ventrorostral subregions of PFC. However, it remains unclear how prefrontal representations could be dissected in a multivariate fashion to reveal (1) what information is encoded in each subregion, (2) whether information systematically transforms across contiguous PFC subregions as a gradient, (3) how this transformation is affected by functional connectivity. Here we shed light on these issues by using RSA to decode informational composition in the PFC while using participant-specific localisers to facilitate individually-tailored precision. Our findings elucidate the functional organisation of PFC by revealing how a trade-off between dimensionality and generalisability unfolds in the PFC and highlight the strength of RSA in deciphering the coding of cognitive control.

neuroscience↗